Avai lable onl ine at www.sc iencedirect .com

V A L U E I N H E A L T H R E G I O N A L I S S U E S 4 C ( 2 0 1 4 ) 1 – 5

2212-1099$36.00 – s

Published by Elsevie

http://dx.doi.org/10.

Conflict of intere

E-mail: epotapch

* Address corresEconomics,” Office

journal homepage: www.elsevier .com/ locate /vhr i

Social Cost of Substance Abuse in RussiaElena Potapchik, PhD*, Larisa Popovich, PhD

Institute of Health Economics, National Research University “Higher School of Economics,” Moscow, Russia

A B S T R A C T

Objective: To summarize results of studies that estimate the socialcosts of alcohol, tobacco, and illicit drug abuse in Russia. The purposeof these studies was to inform policymakers about the real economicburden of risky behaviors and to provide conditions for evidence-based and well-informed decision making in this area. Methods: Thecost-of-illness method was applied to estimate the social cost ofsubstance abuse. The intangible cost was not included in estimation.A prevalence-based approach was applied to estimate the tangiblecost. For the estimation of direct costs, a top-down method was used.Indirect costs were estimated using two methods: the human capitaland the friction cost. Results: In 2008, the social cost of substanceabuse in Russia comprised 677.2 billion rubles if the friction costmethod is applied and 1965.9 billion rubles if the human capitalmethod is used. The social cost of substance abuse is defined to

ee front matter Copyright & 2014, International S

r Inc.

1016/j.vhri.2014.03.004

st: The authors have indicated that they have no

ik@hse.ru.

pondence to: Elena Potapchik, Institute of Healt504, 4, Bld. 2, Slavyanskaya Pl., Moscow, Russia.

the greatest extent by alcohol consumption, comprising about 45%of the economic burden. Illicit drug use comprises about 30% of theeconomic burden and tobacco consumption 25%. Conclusions: Theresults of economic studies demonstrated that psychoactive sub-stances impose a considerable economic burden on society. Analy-sis of the substance abuse social cost pattern shows that the mainlosses that society bears because of these behavioral risk factors falloutside the health care system and lay in other sectors of theeconomy such as social care, law enforcement, and productivitylosses.

Keywords: alcohol, illicit drugs, social cost, substance abuse, tobacco.

Copyright & 2014, International Society for Pharmacoeconomics andOutcomes Research (ISPOR). Published by Elsevier Inc.

Introduction

Risky lifestyle behaviors pose significant challenges to publichealth and impose on society an enormous burden expressed inepidemiological, economic, and social terms. Alcohol, tobacco,and illicit drugs are the main behavioral risk factors [1]. Country-specific studies devoted to the evaluation of the economic impactof substance abuse were carried out in many countries, butmainly in high-income ones. The results of the studies havedemonstrated a significant drain on a country’s economy in bothbudgetary expenditure terms and reduced productivity due tomorbidity and premature mortality.

The rates of psychoactive substance use in Russia are amongthe highest in the world. But there are no country-specificsystematic evaluations of the overall burden imposed on theRussian economy by substance abuse. Restricted data on differ-ent parts of the aggregate burden are published, mainly data onthe epidemiological burden. It is important to have adequatesocial cost estimates to argue that policies to prevent and treatsubstance abuse should be among the highest priorities of thepublic policy agenda. Such estimates would give a broader pictureof the problem looking at it not only from the health sectorperspective but also from that of the economy as a whole.

The article summarizes the results of economic studies thatexamine the effect of substance abuse on the Russian economy

in 2008 (for alcohol and illicit drug abuse) and 2010 (for tobaccoconsumption). For comparability of all three studies, the eco-nomic burden of tobacco consumption initially evaluated for 2010was recalculated in 2008 prices using the consumer price index[2]. The objective was to provide reliable and credible estimates ofthe economic consequences of alcohol, tobacco, and illicit druguse in Russia, which are evaluated in accordance with interna-tionally adopted approaches.

Methods

Study Design

Estimates of the economic burden of a disease can be made usingthree main approaches: the cost-of-illness method, economicgrowth models, and the full income approach [3,4]. The Interna-tional Guidelines for Estimating the Costs of Substance Abuserecommend using the cost-of-illness method, which was appliedin our studies [5]. This method enables the estimation of thetangible and intangible costs of a disease. The intangible cost,which reflects the cost of suffering and pain of substance usersand others as well as the cost of lost life, was not considered inour studies.

ociety for Pharmacoeconomics and Outcomes Research (ISPOR).

conflicts of interest with regard to the content of this article.h Economics, National Research University “Higher School of

Table 1 – Types of social cost included into estimations, by type of substance abuse.

Types of cost Alcohol Illicit drugs Tobacco

Direct medical cost þ þ þDirect nonmedical cost including the following: þ þ þLaw enforcement and the criminal justice þ þ þResearch, public education, and prevention þ þ þFires þ � þRoad accidents þ þServices for orphans þ � �

Indirect cost including the following: þ þ þPremature deaths þ þ þShort-term disability due to illness þ þ þ

Note. “þ” type of cost included, and “�” type of cost not included.

V A L U E I N H E A L T H R E G I O N A L I S S U E S 4 C ( 2 0 1 4 ) 1 – 52

For estimating the tangible cost, the prevalence-basedapproach was applied. The tangible cost included direct andindirect costs.

Direct costs included the following main cost types: medicalcost, law enforcement and the criminal justice costs, cost ofresearch and prevention, and costs of fires, road accidents, andservices for orphans. For estimation of direct costs, the top-townmethod was used. Direct medical cost included public expendi-tures on the provision of medical care for substance users andpeople injured by their actions, for diseases of direct substanceetiology, and for substance-related diseases. Attributable frac-tions were used for determining the number of treated patientswho received medical care for substance abuse–related condi-tions. Morbidity rates were evaluated on the basis of diseaseprevalence rates. Direct nonmedical costs were estimated usingattributable fractions either evaluated directly on the basis ofavailable Russian data and a set of assumptions or as reported incost studies carried out in other countries.

Indirect cost included productivity losses in the workplaceassociated with premature death and short-term disability due toillness. Costs included into estimations by type of substanceabuse are presented in Table 1. Indirect costs were estimated byusing two methods: the human capital method and the frictioncost method. Indirect costs were estimated in terms of the lostearnings stream of individuals. Productivity losses due to pre-mature mortality were calculated on the basis of average age-and sex-specific earnings for the lost years.

Table 2 – Social costs of substance abuse in Russia, 2008

Types of cost Alcohol, 2008

Direct cost including thefollowing:

284,501.6

Medical cost 161,980.2Nonmedical cost including

the following:122,521.4

Law enforcement and thecriminal justice

94,255.6

Research, publiceducation, and prevention

258.8

Fires 17,786.6Road accidents 3,000.3Services for orphans 7,220.1

Friction costmethod

Humancapitalmethod

F

Indirect cost 27,217.6 587,120.9Social cost 311,719.2 871,622.5 2

Social cost can be estimated in terms of gross and net cost. Thefirst considers only negative effects; the second takes into accountany possible positive effects. Among the most pernicious behav-ioral risk factors (alcohol, tobacco, and illicit drugs), only alcohol ischaracterized by having some positive effects on health. Asdemonstrated by international research, a moderate consumptionof alcohol in older ages has a preventive effect, leading to adecreased probability of cardiovascular diseases. That is why whenestimating the social cost of alcohol consumption, the positiveeffects in terms of decrease in hospitalizations should be counted.Some studies that counted for both gross and net social cost ofalcohol consumption demonstrate that the net social cost could belower than the gross social cost by 12.5% to 14.5% [6,7]. In our study,only the gross social cost of alcohol consumption was estimated,assuming that positive effects in Russia could be insignificant dueto much harder structure of alcohol consumption compared withthat in other countries. This assumption is indirectly confirmed bythe data presented in the research on the global burden of diseaseand injury and the economic cost attributable to alcohol use andalcohol-use disorders. In this research, the positive effect of alcoholconsumption for Russia in terms of disability-adjusted life-yearswas evaluated as zero [8].

Data Collection

Our studies mainly relied on Russian official statistics andsurveys carried out in Russia such as data of the Ministry of

(million rubles).

Illicit drugs, 2008 Tobacco, 2008

189,351.3 131,660.1

19,910.5 109,686.8169,440.8 21,973.3

168,374.7 108.1

151.7 141.6

– 21,723.6914.4 –

– –

riction costmethod

Humancapitalmethod

Friction costmethod

Humancapitalmethod

22,222.9 381,823.7 22,290.3 391,468.811,574.2 571,175.0 153,950.4 523,128.9

0%

20%

40%

60%

80%

100%

Alcohol Illicit drugs Tobacco

91.3% 89.5% 85.5%

8.7% 10.5% 14.5%

Friction cost method

Indirect costs

Direct costs

Shar

e of

soci

al c

ost

0%

20%

40%

60%

80%

100%

Alcohol Illicit drugs

Tobacco

32.6% 33.2% 25.2%

67.4% 66.8% 74.8%

Human capital method

Indirect costs

Direct costs

Shar

e of

soci

al c

ost

Fig. 1 – The pattern of substance abuse social costs by main category, friction cost, and human capital methods.

V A L U E I N H E A L T H R E G I O N A L I S S U E S 4 C ( 2 0 1 4 ) 1 – 5 3

Health, the Federal Compulsory Health Insurance Fund, theFederal Treasury, the Ministry of Internal Affairs, the Ministryof Emergency Situations, and the Federal Statistical Service. Datafrom two representative surveys—the Russian Longitudinal Mon-itoring Survey and the Global Adult Tobacco Survey, RussianFederation, 2009—were used.

The main challenge in data supply was in the area ofepidemiological data. There are not many published Russianepidemiological studies devoted to defining the effect of riskybehavioral factors on morbidity and mortality. When data fromthe Russian studies were not available, World Health Organiza-tion data on attributable fractions for Russia were used. Only in acase when no data reflecting Russian realities were available,data of similar studies carried out in other countries were used.

16%

13%

2%

0%

2%

4%

6%

8%

10%

12%

14%

16%

18%

Alcohol Tobacco Illicit drugs

Sha

re o

f pub

lic h

ealth

exp

endi

ture

s

Fig. 2 – Share of public health expenditures spent on thetreatment of patients with substance abuse (%).

Results

In 2008, the social cost of alcohol abuse comprised 311.7 billionrubles if the friction cost method is applied and 871.6 billionrubles if the human capital method is applied. The social cost ofillicit drug abuse comprised 211.6 and 571.2 billion rubles,respectively. The social cost of tobacco comprised 154.0 and523.1 billion rubles, respectively. The results of the estimatesare presented in Table 2.

As can be seen from the above-presented figures, methodsused for the evaluation of indirect costs can significantly affectthe results of social cost estimates. In all considered studies,social costs calculated with the friction cost method are about 3times lower than the same costs calculated when the humancapital method is applied. This is explained by the economicapproaches to the evaluation of productivity losses that underliethese methods.

Furthermore, the application of different methods could leadto changes in economic burden patterns. When substance abusesocial costs are calculated with the use of the same method, thesocial cost pattern by main category is similar for all types ofsubstance abuse (Fig. 1). If the friction cost method is applied, thebulk of the social cost is accounted for by direct cost (about 90% ofaggregate costs for each risky behavior). If the human capitalmethod is applied, the pattern is just the opposite—the bulk ofthe social cost is accounted for by indirect cost (about 70% ofaggregate costs for each risky behavior).

At the same time from the point of view of direct cost pattern,significant differences are observed. The share of medical cost intotal direct cost varies substantially among different types of

substance abuse. For illicit drug abuse, the medical cost sharecomprises only 10.5% of total direct cost; for alcohol and tobaccouse, this indicator increases correspondingly up to 56.9% and 83.3%.The state spends a substantial part of public health expenditures onmedical care for patients with substance abuse. More than one thirdof public funds are spent on the treatment of such patients. Thebiggest share goes to the treatment of alcohol-dependent patientsand the smallest to drug-dependent patients (Fig. 2).

Significant differences in direct nonmedical cost pattern arealso observed because of the numerous and varied set of adverseconsequences involved.

All three psychoactive substances are linked with crime.Tobacco is associated with crime to a much lesser extent than areothers. Tobacco is mainly associated with illegal commodity turn-over, such as smuggling, production, and sale of unbrandedproducts. Alcohol and illicit drugs are associated much more thantobacco use with other types of crime. Alcohol overconsumptionoften leads to burglary, thefts, and crimes of violence and, espe-cially in Russia, road traffic accidents with fatal consequences. In allcountries, except for light narcotics in a few countries, the use ofnarcotics is illegal. Narcotics due to their illegal status are associ-ated with assault, homicide, and other crimes of violence resultingfrom “wars” in the illicit drug market. Tobacco use is associatedmuch more strongly with fire risk than are other substances.

Differences in the consequences of psychoactive substanceuse are reflected in the patterns of their direct nonmedical costs.Almost all the direct nonmedical cost of illicit drug use is

Table 3 – Direct nonmedical cost by type of substance abuse.

Type of directnonmedical cost

Alcohol Illicit drugs Tobacco

Totalcost

(millionrubles)

Attributablefractions (%)

Cost associatedwith substanceabuse (million

rubles)

Totalcost

(millionrubles)

Attributablefractions (%)

Cost associatedwith substanceabuse (million

rubles)

Totalcost

(millionrubles)

Attributablefractions (%)

Cost associatedwith substanceabuse (million

rubles)

Research, publiceducation, andpreventionincluding thefollowing:

414.4 62.46 258.8 257.6 58.89 151.7 141.6

Federal targetedprograms

414.4 62.46 258.8 257.6 58.89 151.7 310.9 16.02 49.8

National PriorityProject Health

306 30 91.8

Law enforcementand criminaljustice includingthe following:

94,255.6 168,374.7 108.1

Law enforcementagencies andjudicial system

665,893.2 13.5 89,895.6 665,893.2 22.81 151,862.2 665,893.2 0.016 108.1

Maintenance ofmedicaldepartments ofsobriety

4,360.0 100 4,360.0

Federal DrugControl Service

165,12.5 100 165,12.5

Fires 88,933.1 20 17,786.6 88,933.1 24.43 21,723.6Road accidents 48,125.8 6.23 3,000.3 48,125.8 1.90 914.4Services for

orphans18,050.25 40 7,220.1

Total directnonmedical cost

122,521.4 169,440.8 21,973.3

VALUE

INH

EALTH

REGIO

NAL

ISSUES

4C

(2014)1–5

4

V A L U E I N H E A L T H R E G I O N A L I S S U E S 4 C ( 2 0 1 4 ) 1 – 5 5

assigned to law enforcement and criminal justice costs, includingthe cost of the Federal Drug Control Service. For alcohol, theshare of this type of cost is lower, but still substantial. The lowestshare of this type of direct nonmedical cost is observed intobacco. The bulk of the direct nonmedical cost of tobacco isassigned to the cost of fires (Table 3).

Discussion

In our cost studies, two methods were applied to estimateindirect cost. This gave an opportunity to present minimal andprobably midlevel evaluations of the costs. Application of theother often used method, willingness to pay, as cost studies ofother countries demonstrate, most likely would give the highestlevel of social cost estimations.

Despite the fact that all our cost studies were carried out inaccordance with internationally adopted methodologicalapproaches, the estimates presented here should be consideredwith some caution.

It is important to emphasize that for several reasons the socialcost estimates presented here could be interpreted as lowerestimates of total societal losses associated with substanceabuse. Because of data limitations, some types of social costwere not included in the estimates such as intangible cost as wellas some types of direct and indirect tangible cost (e.g., householdproduction activities and nursing home costs). Not many coststudies of substance abuse include the intangible cost in theestimations. In the studies in which intangible cost evaluationswere included, however, the total social cost increases substan-tially. For instance, intangible cost comprises 25% of the totalsocial cost of alcohol abuse in Europe [9]. A study by Australianresearchers demonstrates that the share of intangible cost couldbe much higher, and for some types of substance abuse couldeven exceed the tangible cost [10]. As shown by their study, thesocial cost pattern by these categories varies substantially bytypes of psychoactive substances. The lowest share of intangiblecost comprises 15.6% for illicit drug abuse; for alcohol over-consumption, it comprises 29.3%; and the highest share of61.8% is demonstrated for tobacco consumption. Direct extrap-olation of such proportion for Russian estimates is impossibledue to differences in psychoactive substance consumption levels,labor market functioning, life expectancy, and other social,demographic, and economic factors. Thus, estimates of socialcost in Russia presented in the article represent a substantial partof the aggregate social and economic losses associated withsubstance abuse, but still they are only part of those losses.

The social cost of illicit drug abuse is particularly underesti-mated. Unlike statistics on the consumption of other types ofpsychoactive substance, statistics on the prevalence of illicit drugconsumption are less reliable. It is impossible to evaluate the realprevalence rate because of the illegal status of narcotics in allcountries; a considerable part of drug users will be always latent.

The biggest challenge of the studies arose from the lack ofinformation on adequate attributable fractions used for estimat-ing direct and indirect tangible costs. Often, estimates of sub-stance abuse attributable fractions for mortality and morbiditymade for Russia varied considerably. In our studies where datafor Russia were available, conservative estimates were used,which could lead to underestimation. When domestic studieswere not available, attributable fractions used in foreign studieswere applied to our estimates. To a certain extent, this affectedthe reliability of economic evaluations. For example, there are nopublished Russian studies on illicit drug attributable fractions forparticular categories of crime. The attributable fractions publishedin an American study were used in our estimates [11]. The adoptionof such data to Russian realities should be treated with caution

because of differences in legal definitions and crime classifications.Such estimates, however, show initial approximations of costs ofcrime associated with illicit drug abuse. Disregarding such costsbecause of a lack of domestic data could be considered as a biggermistake assuming that such costs do not exist.

In spite of all these reservations, the results of our cost studiescould be considered as the first attempt to provide complexrealistic estimates of social and economic losses associated withsubstance abuse in Russia, which was made in accordance withinternationally adopted methodology and based on available data.

Conclusions

The substance abuse social cost estimates presented here shouldbe considered as underestimates because not all types of costwere included. For example, intangible cost was not included inthe estimates.

The studies considered in the article reveal that the statedevotes tremendous resources to managing diverse effects ofsubstance abuse through providing medical and social care,foster care, incarceration, and various other services. About onethird of the public health expenditure is spent on treatingsubstance abusers. The studies demonstrate that direct medicalcosts are not the main part of the economic burden of substanceabuse. Analysis of the substance abuse social cost pattern givesevidence that the main losses that society bears because of thesebehavioral risk factors fall outside the health care system and layin other areas of economy such as social care, law enforcement,and productivity losses. The study’s results show that the spreadof considered risk factors substantially affects the country’sproductive capacity.

Source of financial support: These findings are the result ofwork supported by Pfizer H.C.P. Corporation and Johnson&Johnson. The views expressed in this article are those of theauthors, and no official endorsement by supported companies isintended or should be inferred.

R E F E R E N C E S

[1] Lopez A, Mathers C, Ezzati M, et al. Global Burden of Disease and RiskFactors. The International Bank for Reconstruction and Development,Washington, DC: The World Bank, 2006.

[2] (Russia in figures 2012). In: Statistical Handbook. Moscow, Russia: TheFederal State Statistics Service (Rosstat), 2012.

[3] WHO. Guide to Identifying the Economic Consequences of Disease andInjury. Geneva, Switzerland: Department of Health Systems Financing,Health Systems and Services, World Health Organization, 2009.

[4] Bloom DE, Cafiero ET, Jane-Llopis E, et al. The Global Economic Burdenof Noncommunicable Diseases. Geneva, Switzerland: World EconomicForum, 2011.

[5] Single E, Collins D, Easton B, et al. International Guidelines forEstimating the Costs of Substance Abuse. (2nd ed.). Ottawa, Canada:Canadian Centre on Substance Abuse, 2001.

[6] Jarl J, Johansson P, Eriksson A, et al. The social cost of alcoholconsumption: an estimation of the economic and human costincluding health effects in Sweden 2002. Eur J Health Econ2008;9:351–60.

[7] Konnopka A, Konig H. Direct and indirect costs attributable to alcoholconsumption in Germany. Pharmacoeconomics 2007;25:606–18.

[8] Rehm J, Mathers C, Popova S, et al. Global burden of disease and injuryand economic cost attributable to alcohol use and alcohol-usedisorders. Ser Alcohol Global Health 2009:373: page range.

[9] Anderson P, Baumberg B. Alcohol in Europe: A Public HealthPerspective. London: Institute of Alcohol Studies, 2006.

[10] Collins DJ, Lapsley HM. The costs of tobacco, alcohol and illicit drugabuse to Australian society in 2004/2005. Short summary, MonographSeries, No. 66 2008.

[11] Office of National Drug Control Policy. The Economic Costs of DrugAbuse in the United States, 1992-2002 (Publication No. 207303).Washington, DC: Executive Office of the President, 2004.